31 Oct 2025 • The Analysis

Optis v Apple: Can a FRAND price be inferred from "comparable" SEP licences that are not FRAND?

4 minute read

Share

In May this year, the Court of Appeal (England and Wales) determined that Apple must pay Optis $502 million for an 11-year global licence covering its patents essential to mobile-telecommunication standards – almost ten times the $56 million awarded at first instance.[1] In this article, Pekka Sääskilahti, Andrew Tuffin and Timo Autio [2] look at the reasons for that dramatic increase, and draw out lessons for future FRAND (fair, reasonable and non-discriminatory) determinations.

Read the full article here

Both Optis v Apple judgments sought to determine a FRAND price by reference to “comparable” contracts – similar licences that Optis and Apple had agreed with third parties. Optis identified 9; Apple identified 14. By the time of the Appeal, 5 remained: 1 Optis licence, and 4 Apple licences. However, the judgments used different methodologies to (a) unpack the prices agreed for those contracts and (b) evaluate them – to determine how closely comparable the contracts are, and how reasonable and reliable their unpacked prices are. The gap between the two determinations demonstrates how much the specifics of a methodology matter at each stage (Figure 1).

The approach the original decision took at each stage was appealed on several grounds. Many were upheld. In its place, the Appeal infers a FRAND price from the comparable contracts, but it was not determined by those contracts directly nor by taking an average. This is significant. The judgment finds that the range of prices agreed for the parties’ comparables was too wide to indicate a plausible FRAND range – the contracts were either not comparable, or their prices were not reasonable or reliable.

Even so, those contracts were still useful. Comparability, reliability, and reasonableness are matters of degree. A court can ask: How closely comparable is each contract to the disputed licence? How much uncertainty affects the reliability of its unpacked price? To what extent is the price agreed inflated by hold up or depressed by hold out? It can then take all of that into account.

That is essentially what the Court of Appeal did: it found that Apple’s licences implied prices that were too low to be FRAND, and Optis’s implied a price that was too high; and from there, having considered the degree of distortion, it determined the FRAND rate.

Drawing on that approach, we set out six lessons to assist courts in analysing comparables in future.

1. Presenting the evidence: “unpacking” the prices that parties agreed

  • Lesson 1: The distinction between reliable and comparable prices. Unpacking expresses the prices that parties actually agreed in a common “currency”. It should not adjust those prices to estimate what the parties might have agreed for more comparable contracts, nor should it prescribe a payment structure that comparable contracts must have.
  • Lesson 2: The benefit of using both parties’ comparables. Where reliable, mapping both the prices of the licensor’s and the licensee’s comparables onto a common scale — the implied price for the overall SEP “stack” — helps reduce the risk of relying on contracts with a common bias.

2. Evaluating the evidence: assessing how closely comparable and reasonably priced those contracts are

  • Lesson 3: Analysing the range of unpacked rates. A wide range of unpacked prices cannot all be comparable, FRAND, and reliable. The wider the range, the less likely it is that all prices are closely comparable, FRAND, and reliable.
  • Lesson 4: Analysing how closely comparable the contracts are. Comparability is a matter of degree. The closest comparables to the disputed licence are the most informative of the price parties would agree for it.
  • Lesson 5: Analysing how reasonable the prices of the closest comparable are. Even the closest comparables may not be FRAND. In this case, all five of the comparables were affected by some degree of hold up or hold out.
  • Lesson 6: Inferring the FRAND rate. Even non-FRAND contracts may help establish the FRAND rate, provided that (a) there is a structured assessment of how closely comparable they are, and (b) there is a reliable cross-check to assess the extent to which the implied price is reasonable.
Figure 1: Illustration of 4G SEP rates for smartphones assessed in Optis v Apple
Source: Compass Lexecon analysis based on data from [2025] EWCA Civ 552, [2017] EWHC 711 (Pat), [2019] TCL Communication Technology Holdings Ltd v Telefonaktiebolaget LM Ericsson, USCA Fed. Cir. No. 18-1363, and a translation of [2023] OPPO v Nokia, The People’s Republic of China Chongqing First Intermediate Court (2021 Yu 01 Minchu no. 1232), page 100 (‘Reasoning’ Section V. (I) 1.)). Note: Implied Aggregate Royalty Rate (ARR) as % of a representative $470 phone..

References

  1. For the Appeal, see [2025] EWCA Civ 552. For the original decision, see [2023] EWHC 1095 (Ch). Apple sought permission to appeal to the UK Supreme Court against the orders giving effect to the Court of Appeal’s ruling. UKSC/2025/0145. https://www.supremecourt.uk/ca... Updated 6 November: On 4 November, the UK Supreme Court granted Apple permission to appeal.

  1. Pekka Sääskilahti is an Executive Vice President, Andrew Tuffin is a Vice President, and Timo Autio is a Vice President at Compass Lexecon. We have benefitted from comments by Ben Dubowitz and Janne Sirnio. Adam Sanderson provided research assistance. The views expressed in this article are the views of the authors only and do not necessarily represent the views of Compass Lexecon, its management, its subsidiaries, its affiliates, its employees or its clients.

A new version of Compass Lexecon is available.